Sparse pseudo-input local Kriging for large spatial datasets with exogenous variables. (3rd March 2020)

• Record Type: Journal Article
• Authors: Farmanesh, Babak; Pourhabib, Arash
• Is Part Of: IISE transactions. Volume 52:Number 3(2020)
• Page Start: 334
• Page End: 348
• Publication Date: 2020-03-03
• Subjects: Gaussian process regression -- local Kriging -- sparse approximation -- spatial datasets; Industrial engineering -- Periodicals; Systems engineering -- Periodicals; Industrial engineering; Systems engineering; Electronic journals; Periodicals; 670.285
• DOI: 10.1080/24725854.2019.1624926
• Languages: English
• ISSNs: 2472-5854

Abstract

Abstract: We study large-scale spatial systems that contain exogenous variables, e.g., environmental factors that are significant predictors in spatial processes. Building predictive models for such processes is challenging, due to the large numbers of observations present making it inefficient to apply full Kriging. In order to reduce computational complexity, this article proposes Sparse Pseudo-input Local Kriging (SPLK), which utilizes hyperplanes to partition a domain into smaller subdomains and then applies a sparse approximation of the full Kriging to each subdomain. We also develop an optimization procedure to find the desired hyperplanes. To alleviate the problem of discontinuity in the global predictor, we impose continuity constraints on the boundaries of the neighboring subdomains. Furthermore, partitioning the domain into smaller subdomains makes it possible to use different parameter values for the covariance function in each region and, therefore, the heterogeneity in the data structure can be effectively captured. Numerical experiments demonstrate that SPLK outperforms, or is comparable to, the algorithms commonly applied to spatial datasets.